Easymode: general pretrained networks for cellular cryo-ET enable flexible approaches to subtomogram averaging
The paper introduces Easymode, a library of pretrained general segmentation networks trained on over 4,000 tilt series that enables flexible, training-free workflows for identifying, extracting, and analyzing macromolecules in cellular cryo-ET, demonstrated by determining the 4.0 Å in situ structure of IMPDH filaments and mapping their cellular environment.